Miyamoto, Masayuki

Objective To apply a machine learning analysis to clinical and presynaptic dopaminergic imaging data of patients with rapid eye movement (REM) sleep behavior disorder (RBD) to predict the development of Parkinson disease (PD) and dementia with Lewy bodies (DLB). Method s In this multicenter study of the International RBD study group, 173 patients (mean age 70.5 ± 6.3 years, 70.5% males) with polysomnography‐confirmed RBD who eventually phenoconverted to overt alpha‐synucleinopathy (RBD due to synucleinopathy) were enrolled, and underwent baseline presynaptic dopaminergic imaging and clinical assessment, including motor, cognitive, olfaction, and constipation evaluation. For comparison, 232 RBD non‐phenoconvertor patients (67.6 ± 7.1 years, 78.4% males) and 160 controls (68.2 ± 7.2 years, 53.1% males) were enrolled. Imaging and clinical features were analyzed by machine learning to determine predictors of phenoconversion. Results Machine learning analysis showed that clinical data alone poorly predicted phenoconversion. Presynaptic dopaminergic imaging significantly improved the prediction, especially in combination with clinical data, with 77% sensitivity and 85% specificity in differentiating RBD due to synucleinopathy from non phenoconverted RBD patients, and 85% sensitivity and 86% specificity in discriminating PD‐converters from DLB‐converters. Quantification of presynaptic dopaminergic imaging showed that an empirical z ‐score cutoff of −1.0 at the most affected hemisphere putamen characterized RBD due to synucleinopathy patients, while a cutoff of −1.0 at the most affected hemisphere putamen/caudate ratio characterized PD‐converters. Interpretation Clinical data alone poorly predicted phenoconversion in RBD due to synucleinopathy patients. Conversely, presynaptic dopaminergic imaging allows a good prediction of forthcoming phenoconversion diagnosis. This finding may be used in designing future disease‐modifying trials. ANN NEUROL 2024;95:1178–1192

Abstract This is an international multicentre study aimed at evaluating the combined value of dopaminergic neuroimaging and clinical features in predicting future phenoconversion of idiopathic REM sleep behaviour (iRBD) subjects to overt synucleinopathy. Nine centres sent 123I-FP-CIT-SPECT data of 344 iRBD patients and 256 controls for centralized analysis. 123I-FP-CIT-SPECT images were semiquantified using DaTQUANTTM, obtaining putamen and caudate specific to non-displaceable binding ratios (SBRs). The following clinical variables were also analysed: (i) Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale, motor section score; (ii) Mini-Mental State Examination score; (iii) constipation; and (iv) hyposmia. Kaplan-Meier survival analysis was performed to estimate conversion risk. Hazard ratios for each variable were calculated with Cox regression. A generalized logistic regression model was applied to identify the best combination of risk factors. Bayesian classifier was used to identify the baseline features predicting phenoconversion to parkinsonism or dementia. After quality check of the data, 263 iRBD patients (67.6 ± 7.3 years, 229 males) and 243 control subjects (67.2 ± 10.1 years, 110 males) were analysed. Fifty-two (20%) patients developed a synucleinopathy after average follow-up of 2 years. The best combination of risk factors was putamen dopaminergic dysfunction of the most affected hemisphere on imaging, defined as the lower value between either putamina (P < 0.000001), constipation, (P < 0.000001) and age over 70 years (P = 0.0002). Combined features obtained from the generalized logistic regression achieved a hazard ratio of 5.71 (95% confidence interval 2.85–11.43). Bayesian classifier suggested that patients with higher Mini-Mental State Examination score and lower caudate SBR asymmetry were more likely to develop parkinsonism, while patients with the opposite pattern were more likely to develop dementia. This study shows that iRBD patients older than 70 with constipation and reduced nigro-putaminal dopaminergic function are at high risk of short-term phenoconversion to an overt synucleinopathy, providing an effective stratification approach for future neuroprotective trials. Moreover, we provide cut-off values for the significant predictors of phenoconversion to be used in single subjects.